1. Field of the Invention
The present invention relates to a picture reproduction apparatus and a picture decoding apparatus used for reproducing a digital compressed video image recorded on a recording medium.
2. Description of the Related Art
An apparatus for recording/reproducing a compressed digital video image has been developed with the progress of picture compression techniques. FIG. 18 is a block diagram of a conventional picture recording/reproducing apparatus. The conventional picture recording/reproducing apparatus shown in FIG. 18 includes a picture encoding section 91 using an intra-frame discrete cosine transform (DCT) encoder, a recording section 92, a recording medium 93, a reproducing section 94, and a picture decoding section 95 using an intra-frame DCT decoder. A digital video image taken by a camera or the like is input into the intra-frame DCT encoder 91. In the intra-frame DCT encoder 91, the digital video image is encoded by data processing such as DCT and variable length encoding, frame by frame. Then, the processed video signal is rearranged into a recording format, and an error correction code is added thereto in the recording section 92. Thereafter, the reformatted digital video signal with the error correction code is recorded onto the recording medium 93. During the playback, any error in the reproduced signal is corrected, and the reproduced signal is reformatted into a decoding format in the reproducing section 94. Thereafter, the reproduced signal is decoded by a data process such as variable length decoding and inverse DCT in the picture decoding section 95, so as to obtain a reproduced video image. As to the intra-frame encoding, a usual encoding is completed in a data unit of one frame or one block consisting of (8.times.8) pixels or the like. Accordingly, during playback in fast forward or fast reverse in which only a part of data can be reproduced, the resulting video image has a certain quality with which the contents of the image can be perceived (see, for example, "Study on Image Quality Reproduced at Variable Speeds by Digital VCR", by Takakura et al., Extended Abstracts; The Annual Meeting of The Institute of Television Engineers, 1992, pp. 93-94).
FIG. 19 is a block diagram of a conventional picture decoding apparatus using motion-compensation and inter-frame encoding techniques. In FIG. 19, a picture decoding section 4 includes a buffer 41, a variable length decoder 42, an inverse quantization circuit 43, an inverse DCT circuit 44, and a motion compensating circuit 45. Also, a frame memory 5 is provided. First, compressed image data is input into the buffer 41. Then, the data is decoded by processing of variable length decoding, inverse quantization, and inverse DCT, so as to obtain real-time data. In this scheme, the inter-frame decoding technique is used, so that the difference between successive frames is encoded. Accordingly, when the image data is to be recovered, the decoded difference data is added to a block obtained by motion-compensating the data of previous frames stored in the frame memory 5 by each block, so as to obtain the decoded image data.
According to the inter-frame encoding by which the difference is sent, a motion component is effectively transmitted by motion compensation. Thus, the inter-frame encoding system can attain high data compression rates while suppressing deterioration of image quality, as compared with the intra-frame encoding system shown in FIG. 18 (see, for example, ISO/IEC/DIS 11172 "Information Technology--Coding of moving pictures and associated audio for digital storage media up to about 1.5 Mbps", 1992, p. 61).
Since the system shown in FIG. 18 uses the intra-frame encoder, there is a disadvantage in that the data compressibility can not be set as high as that in the inter-frame encoding system or the like. This disadvantage can be overcome by using an inter-frame encoding technique in the picture encoding section and decoding section as in the system shown in FIG. 19. However, the difference data is encoded, so that the current frame can be reproduced only after the image of previous frame is reproduced. Thus, the system shown in FIG. 19 also has a disadvantage in that the desired image cannot be obtained during playback in fast forward or fast reverse in which only a part of the image data can be reproduced.
In slow playback at a half speed or the like, as shown in FIG. 20, the output from the reproducing section includes mixed reproduced image data and other invalid data (cross talk, noise etc.) in a doubled period as compared with normal playback. The kind of invalid data is not assured, so that a synchronization pattern or the like may possibly be output. Therefore, if the reproduced output is directly sent to the decoding apparatus, malfunctions of the decoding apparatus such as turbulence in the reproduced image occurs due to the invalid data portion.